MLOPS=ML+DEVOPS

This was really learnable experience doing this task during my training as LinuxWorld Informatics Pvt Ltd given by our mentor, guide and teacher Vimal Daga sir. In this task we needed to integrate machine learning with devops. We have train a model which could be found on my github https://github.com/Kritika-S/mypro and find its accuracy and integrate it with jenkins to automate it.

Task Overview:

1. Create a container image that has Python3 and Keras, numpy, tensorflow installed using Dockerfile.

2. When we launch this image, it should automatically starts train the model in the container.

3. Create a job chain of job1, job2, job3 and job4 using build pipeline plugin in Jenkins

 4. Job1 : Pull the Github repo automatically when some developers push repo to Github.

5. Job2 : Jenkins should automatically start the respective machine learning software installed interpreter install image container to deploy code and start training.

6. Job3 : After training the model we predict the accuracy or metrics. If metrics accuracy is less than 95%, then tweak the machine learning model architecture. Retrain the model and get the train model.

7. Job4: This job is executed after the success of job3 and we know that the desired accuracy have been obtained.

Dockerfile image created:

Adding Webhooks to repository

Uploading code to github using git.

Creating job chains

Creating a Build pipeline of jobs, we can see the job chain, how they are executed and how the are successfully run.